Arca Marjorie J, Rangel Shawn J, Hall Matt, Rothstein David H, Blakely Martin L, Minneci Peter C, Shah Samir S, Heiss Kurt F, Snyder Charles L, Berman Loren, Browne Marybeth, Vinocur Charles D, Raval Mehul V, Goldin Adam B
*Medical College of Wisconsin/Children's Hospital of Wisconsin, Milwaukee, WI †Boston Children's Hospital and Harvard Medical School, Boston, MA ‡Children's Hospital Association, Overland Park, KS §Women and Children's Hospital of Buffalo, Buffalo, NY ||Monroe Carell Jr. Children's Hospital/Vanderbilt University Medical Center, Nashville, TN ¶The Research Institute at Nationwide Children's Hospital, Columbus #Cincinnati Children's Hospital Medical Center, Cincinnati, OH **Emory University, Children's Healthcare of Atlanta, Atlanta, GA ††Children's Mercy Hospital, Kansas City, MO ‡‡Nemours Alfred I. duPont Hospital for Children, Wilmington, DE §§Lehigh Valley Physician Group, Allentown, PA ||||Seattle Children's Hospital, Seattle, WA.
J Pediatr Gastroenterol Nutr. 2017 Aug;65(2):232-236. doi: 10.1097/MPG.0000000000001523.
Emergency department (ED) visits and hospital readmissions are common after gastrostomy tube (GT) placement in children. We sought to characterize interhospital variation in revisit rates and explore the association between this outcome and hospital-specific GT case volume.
We conducted a retrospective cohort study from 38 hospitals using the Pediatric Health Information System database. Patients younger than 18 years who had a GT placed in 2010 to 2012 were assessed for a GT-related (mechanical or infectious) ED visit or inpatient readmission at 30 and 90 days after discharge from GT placement. Risk-adjusted rates were calculated using generalized linear mixed-effects models accounting for hospital clustering and relevant demographic and clinical attributes, then compared across hospitals.
A total of 15,642 patients were included. A median of 468 GTs were placed in all the 38 hospitals during 3 years (range: 83-891), with a median of 11.4 GT placed per 1000 discharges (range: 2.4-16.7). Median ED visit for each hospital at 30 days after discharge was 8.2% (range: 3.7%-17.2%) and 14.8% at 90 days (range: 6.3%-26.1%). Median inpatient readmissions for each hospital at 30 days after discharge was 3.5% (range: 0.5%-10.5%) and 5.9% at 90 days (range: 1.0%-18.5%). Hospital-specific GT placement per 1000 discharges (rate of GT placement) was inversely correlated with ED visit rates at 30 (P = 0.007) and 90 days (P = 0.020). The adjusted 30- and 90-day readmission rate and the adjusted 30- and 90-day ED return rates decreased with increasing GT insertion rate (P < 0.001).
Higher hospital GT insertion rates are associated with lower ED revisit rates but not inpatient readmissions.
儿童胃造口管（GT）置入术后急诊就诊和再次入院情况较为常见。我们试图描述医院间再就诊率的差异，并探讨这一结果与医院特定GT病例数量之间的关联。
我们利用儿科健康信息系统数据库对38家医院进行了一项回顾性队列研究。对2010年至2012年期间置入GT的18岁以下患者，在GT置入出院后30天和90天评估与GT相关（机械性或感染性）的急诊就诊或住院再入院情况。使用广义线性混合效应模型计算风险调整率，该模型考虑了医院聚类以及相关的人口统计学和临床特征，然后在各医院之间进行比较。
共纳入15642例患者。38家医院在3年期间置入GT的中位数为468根（范围：83 - 891根），每1000例出院患者中置入GT的中位数为11.4根（范围：2.4 - 16.7根）。各医院出院后30天的急诊就诊中位数为8.2%（范围：3.7% - 17.2%），90天为14.8%（范围：6.3% - 26.1%）。各医院出院后30天的住院再入院中位数为3.5%（范围：0.5% - 10.5%），90天为5.9%（范围：1.0% - 18.5%）。每1000例出院患者的医院特定GT置入率（GT置入率）与30天（P = 0.007）和90天（P = 0.020）的急诊就诊率呈负相关。随着GT置入率的增加，调整后的30天和90天再入院率以及调整后的30天和90天急诊返回率均下降（P < 0.001）。
较高的医院GT置入率与较低的急诊再就诊率相关，但与住院再入院率无关。
